What did we learn from tiering in generation four?

[lati0s]

Since the main argument going around here seems to be what the general player base would think of this, would anyone mind if I put up a poll in uncharted territory to get some opinions on this? (or if someone else wants to put it up that's fine).

Another tangentially related issue that I think needs to be addressed is the philosophy of why we ban pokemon. Do we ban pokemon only because we believe that they are too powerful relative to the other pokemon in the metagame, or do we ban pokemon because we prefer to play in the metagame without them? This is an issue that was very muddied through all of gen 4 tiering. Measures such as mandatory paragraph submissions on the characteristics of uber and the characteristics of uber themselves as well as suspect exp would suggest that our goal was to make sure that things were only voted uber if they were 'overpowered'. Yet we continually had suspect tests where the suspect ladder consisted of the metagame without the suspect being tested. If your goal is to determine if something is overly powerful in a given metagame then playing in the metagame with that pokemon removed does not give you any relevant information. If we were voting to see if something fit the characteristics of uber then all the tests should have been conducted in a similar manner to the first shaymin-s test.

 

[capefeather]

No reason to kill the voice of reason. I definitely don't think we should feel like "Start from Scratch" party is the dominant one, nor that their bullish attitude towards the issue is the true voice of reason here.

Please, let's not go about blindly characterizing each other.

Re: UT

As far as I'm concerned, the UT discussions are largely pointless, anyway. In any case, I'm sure that, if people are so willing to theorymon so early in Gen V, they can use their heads a little and talk about Pidgeot in an NU metagame context, or Dragonite in an OU/UU context, or something.

Re: "Wasting time"

This completely misses the point. The point of a test is to form a solid conclusion about one metagame compared to another. That is, the whole point of a test is to learn something. If learning something is not important, then we might as well make up a static banlist and never question it, instead defining these Pokémon as essentially nonexistent. There's no point in testing controversial suspects; just boot them to Uber. It's all quick and painless, with no regard to whether one Pokémon or another was tiered unfairly. There is no point in having any sort of process to sort this out, because we're deciding each Pokémon's "guilt", rather than determining a statement of fact about it. It's the most practical solution.

I don't mean to characterize the entire "other side" in this manner. It's just this argument in particular that gets me.

The question is simply whether it's worth the fallout to start with no bans. Time shouldn't really be that much of a factor because a timeline of "a few months" (the period that seems to be thrown around the most) is still a lot shorter and much more practical than a timeline of two years. A few months rarely allow Nintendo to introduce much, whereas with Gen IV the next generation was already at our doorstep. The real issue at hand is acceptance by the general competitive community.

 

[cim]

did something happen in the last 24 hours or do we seriously have a metagame where pretty much every pokemon is turned upside down due to completely new ability and movepool distribution

how can you theory ban with such a dramatically clean slate?

 

[Firestorm]

Wouldn't starting our tier lists with the ones that Nintendo uses for the Random Wi-fi battles be a good place to start when it comes to making the site accessible to new users? I think that might be a valuable place to start.

This has been brought up multiple times over the past few months as it's the banlist Nintendo uses in Battle Tower, some of their official tournaments, and now on Random WiFi. It keeps it very basic. I would agree with either this or a clean slate with no bans whatsoever, but for some reason very few people give this idea the attention it deserves. I've seen more and more PR / IS members warming up to the idea over this time which is nice.

Honestly though, throwing around the idea amongst people who play Pokemon with Smogon's rules but more casually, I've seen quite a positive response to unbanning everything and starting from scratch.

 

[Seven Deadly Sins]

What's the starting Nintendo banlist?

 

[shrang]

If a Pokemon is obviously broken after some playtesting (Which the 670+ Ubers are likely to be), why don't we just give it a shorter testing period to avoid wasting time?? I believe we should test everything, even when it is likely that it is extremely broken, just to confirm that we are correct. If need be, why don't we just test say Kyogre for 2 days. Then, we can have a general consensus vote if it should be banned immediately, or we should test him further and give him a more detailed test.

 

[Firestorm]

What's the starting Nintendo banlist?

Generation IV:
Mewtwo
Mew
Lugia
Ho-oh
Celebi
Kyogre
Groudon
Rayquaza
Jirachi
Deoxys (all forms)
Dialga
Palkia
Giratina
Phione
Manaphy
Darkrai
Shaymin (all forms)
Arceus
[Item] Soul Dew

Generation V likely includes all of the above + the two Cover Pokemon + Victini.

If a Pokemon is obviously broken after some playtesting (Which the 670+ Ubers are likely to be), why don't we just give it a shorter testing period to avoid wasting time?? I believe we should test everything, even when it is likely that it is extremely broken, just to confirm that we are correct. If need be, why don't we just test say Kyogre for 2 days. Then, we can have a general consensus vote if it should be banned immediately, or we should test him further and give him a more detailed test.

Two days is hardly enough for the dust to settle =/ I don't see the point of going into this predisposed towards a certain outcome. That's no different than banning them from the start. How do we see if something is balanced if we don't even give people time to develop strategies against it?

 

[shrang]

2 days was just an example. If it's too short, we can give it a week. The point was that we don't have to give the Suspect the full length of the test if it's very likely to be broken. At the end of the shortened test period, we can ask the players whether it indeed is overwhelming, if it is, ban it, if it isn't, give it a more detailed test.

 

[lilyhollow]

Since the main argument going around here seems to be what the general player base would think of this, would anyone mind if I put up a poll in uncharted territory to get some opinions on this? (or if someone else wants to put it up that's fine).

I expected others to respond to this, and I don't think you've actually put up a poll yet, but I personally think this would be a good idea. It's not a perfect gauge, but I think it could still provide valuable information.

 

[Seven Deadly Sins]

Generation IV:
Mewtwo
Mew
Lugia
Ho-oh
Celebi
Kyogre
Groudon
Rayquaza
Jirachi
Deoxys (all forms)
Dialga
Palkia
Giratina
Phione
Manaphy
Darkrai
Shaymin (all forms)
Arceus
[Item] Soul Dew

Generation V likely includes all of the above + the two Cover Pokemon + Victini.

Two days is hardly enough for the dust to settle =/ I don't see the point of going into this predisposed towards a certain outcome. That's no different than banning them from the start. How do we see if something is balanced if we don't even give people time to develop strategies against it?

I'd be fairly ok with this if it weren't for the banning of event legendaries like Jirachi, Shaymin, Manaphy, Celebi, Darkrai, etc. I've never been a fan of the arbitrary bans on event Pokemon simply because at the time of release they're not meant to be available, considering that all of these pokemon are available through trading forward from Gen. 4.

 

[Chou Toshio]

Two days is hardly enough for the dust to settle =/ I don't see the point of going into this predisposed towards a certain outcome. That's no different than banning them from the start. How do we see if something is balanced if we don't even give people time to develop strategies against it?

And can you believe that people in this same thread are trying to argue, "don't worry, testing won't take time!" :/

Thank you my friend, you have just pointed out why:

I don't like the idea of not banning Kyogre/Mewtwo/Rayquaza/etc, aka the 'obviously broken' Pokemon. I understand the importance of not making incorrect assumptions about tiering, and I know a lot of new stuff has come up with Black and White. However, I think we are wasting a lot of time in UT right now pretending we won't end up banning these Pokemon.

I don't know.... "lets wait it out and test it and if its broken ok" seems like a really silly thing to do for Pokemon like that when we're talking about wasting months of time. Isn't that what we're trying to avoid?

Continuing on:

P
Re: "Wasting time"

This completely misses the point. The point of a test is to form a solid conclusion about one metagame compared to another. That is, the whole point of a test is to learn something. If learning something is not important,

Chou: Frankly, it's not.

then we might as well make up a static banlist and never question it, instead defining these Pokémon as essentially nonexistent. There's no point in testing controversial suspects; just boot them to Uber. It's all quick and painless, with no regard to whether one Pokémon or another was tiered unfairly. There is no point in having any sort of process to sort this out, because we're deciding each Pokémon's "guilt", rather than determining a statement of fact about it. It's the most practical solution.

Exactly! Delete the bolded part, which in my opinion is irrelevant and contains an irrelevant value-judgement with the word "fair" and you've pretty much summed it up perfectly! Though I would have written this with less negative overtones because I honestly see it as perfectly fine and dandy.

I don't mean to characterize the entire "other side" in this manner.

Don't worry about that-- where's the characterization? You pretty much hit the nail on the head and summarized my whole point without me having to say anything. O.o I don't get this post, because you just restate the other side of the argument and then do not refute it. Frankly, in some ways, I even like your wording/reasoning for my point better than my own.

The question is simply whether it's worth the fallout to start with no bans. Time shouldn't really be that much of a factor because a timeline of "a few months" (the period that seems to be thrown around the most) is still a lot shorter and much more practical than a timeline of two years. A few months rarely allow Nintendo to introduce much, whereas with Gen IV the next generation was already at our doorstep. The real issue at hand is acceptance by the general competitive community.

Just pointing out that just because "A few months" is "shorter and better than what we did before" that still doesn't mean that a few months should be considered reasonable (though I think it would be-- I just don't believe it would actually finish in a few months).

After all, there have been a number of voices in this thread pointing out that testing really had little to no value anyway last gen-- will a shorter process really have any more value? What "facts" are you talking about? It's hard to tell if testing actually shaped people's opinions or not anyway.

Or is tiering all, in fact, arbitrarily based on people's preconceived conclusions no matter what you do?

 

[Chou Toshio]

Edit: There was a post above me, this was not a double-post. :P

Thinking about it more, I really do not get why I am arguing with the "all unban" theory. My only real reason is because I do strongly believe that all bans are pretty arbitrarily based on peoples opinions, and that "testing" is pointless. The only thing that matters is getting to the metagame people want.

And frankly, just by shutting up and letting the all unban happen, it is highly likely the metagame will progress in a fashion that I want-- that being, a fast dynamic metagame played with resource management as a game of attrician, rather than "resistance-based stall circle-jerk" as I have described it not-so-eloquently in past threads. In fact, I have never really supported a ban (I think I might have a few anti-manaphy, anti-latias posts somewhere, but overall I didn't care much), and otherwise was always adamantly arguing against the ban.

I think one thing that can be ascertained by all of this is that there is a general agreement that we want a lot less bans and much more liberal tiering in general at the beginning of 5th gen. The only thing that really comes to the table is the question of time, and the value of actually "testing." Frankly, I think the word "testing" is mis-leading since nothing is actually being measured . . .


That said, there's really nothing wrong with starting from no bans.

All I would personally ask (to take that plan seriously), is that someone outline a strict and concrete plan that outlines the entire process taking place in 4 months, which I would consider a reasonable timeline. Somewhere between 3-5 months, ie. 4 months. Others may have other preferences for the amount of time, but I think we should be as strict as possible, and the pro-unban lobbiers should be strict on themselves with the amount of time. The stricter you are (the less time you ask for), the more likely people will want to support the plan.

Once such a plan is devised, than I think people can actually have real discussion on whether or not we should implement such a plan.

 

[capefeather]

Chou: Frankly, it's not.

This is all there really is to what you said. You didn't have to rub in my face your opposition to my premise -.- But I'm kind of surprised that I'd even have to defend it.

Last time I checked, this was Smogon University, the premiere competitive Pokémon website. Smogon is this way because, rather than making up apparent statements of fact, (Jirachi flinched my Magnezone to death! Uber!) we learn statements of fact about the relevant metagames, and then we pass on what we learn to others. C&C doesn't even let theorymon sets in anymore (that may change with garbage Pokémon analyses in Gen V but that's rather irrelevant); each set is tested by multiple people. In fact, basically every facet of Smogon has playtesting as a cornerstone, including UU, Little Cup and CAP.

Learning through testing is not just important; it's at the core of how Smogon functions. Without learning, we get a bunch of people who have "street smarts" about the game but can't explain what they're playing to anyone else other than vague, one-sentence explanations. It's rather ironic how the very thing that people fear will push people away from Smogon is, in fact, practically essential for what Smogon is supposed to do: pass on knowledge of competitive Pokémon.

---

I'm also rather surprised that you've imposed a four-month limit on testing plans when the closest precedent to it (UU testing) lasted over a year. You're asking us to be idealistic to a fault, when a strict timeline is probably completely unrealistic. However, I did propose some form of a plan earlier (the edited part), but I guess no one saw it, or we were too stuck on whether to do it at all rather than what the plan would be. I'll admit that it isn't that concrete as it is, but I originally wanted feedback before entertaining the idea further.

 

[Seven Deadly Sins]

This is from the thread, and I really wanted to highlight this, as it's the best argument against planning on quickbanning the BST 670+ Pokemon.

I would not say that a no-Ubers start would be sensible, as it would encourage fast, low-deliberation banning for the sake of gameplay that would start with the obvious suspects (Arceus, Kyogre) and quite possibly run down the totem pole too far and overshoot completely balanced metagames.

If we plan on spending an extremely short amount of time on "obvious ubers" then we very heavily run the risk of just carrying on and quickbanning things that may not be broken with the idea that "eventually we'll stop"

Not only do quickbans bias against the top legendaries anyway, but also run the very notable risk of not stopping soon enough. I'd say "ban 670+es off the bat go from there" is probably the best way to go, as even some of the "no" votes in the thread are less based on wanting "no bans" and more about not wanting to overban. I'd say that banning the 670+ Pokemon early prevents us from getting into a "ban everything that looks overpowered" mentality that might end with us throwing the baby out with the bathwater.

 

[lilyhollow]

I agree that we should be concerned with bias against the top legendaries, and with "overshooting." Banning 670+ BST Pokemon right away, though, is a clumsy solution to both of those things. If most people really are not worried about having an OU metagame that "resembles Ubers," which your poll theoretically suggests (though many players may not grasp the potential implications, as far as aesthetics go, of officially making OU the "first balanced metagame"), then it seems like "No Banlist" is the correct decision. I don't think it's asking very much at all that we develop a tiering system that doesn't fall victim to silly biases. New UU seemed to do pretty well for itself anyway.

Right now I do support starting 5th gen OU with no bans, mostly due to the results of the poll. My reasoning hasn't changed, though, and I still don't think that arguments of "giving each Pokemon their fair shot," "being scientific," or "least bans = most competitive" hold any water whatsoever. I think it's a good idea to pander to the surprisingly-numerous people who believe those things, though.

 

[cim]

This is from the thread, and I really wanted to highlight this, as it's the best argument against planning on quickbanning the BST 670+ Pokemon.



If we plan on spending an extremely short amount of time on "obvious ubers" then we very heavily run the risk of just carrying on and quickbanning things that may not be broken with the idea that "eventually we'll stop"

Not really. If we demand a supermajority and have at least 2 weeks to test somehow I doubt that we will have more than 67% of people vote for something like dugtrio to get banned !

really i just don't see smogonites being that rash (i mean people that could actually qualify to vote, not your average poster). when something is first announced and before anyone has played with it, yes maybe. but after two weeks and only taking the top players, i doubt it.

keep in mind most of the quick-to-conclusion backlash from lsat gen came with UNbans like wobb, not bans like salamence

 

[Chou Toshio]

This is all there really is to what you said. You didn't have to rub in my face your opposition to my premise -.- But I'm kind of surprised that I'd even have to defend it.

Last time I checked, this was Smogon University, the premiere competitive Pokémon website. Smogon is this way because, rather than making up apparent statements of fact, (Jirachi flinched my Magnezone to death! Uber!) we learn statements of fact about the relevant metagames, and then we pass on what we learn to others. C&C doesn't even let theorymon sets in anymore (that may change with garbage Pokémon analyses in Gen V but that's rather irrelevant); each set is tested by multiple people. In fact, basically every facet of Smogon has playtesting as a cornerstone, including UU, Little Cup and CAP.

Learning through testing is not just important; it's at the core of how Smogon functions. Without learning, we get a bunch of people who have "street smarts" about the game but can't explain what they're playing to anyone else other than vague, one-sentence explanations. It's rather ironic how the very thing that people fear will push people away from Smogon is, in fact, practically essential for what Smogon is supposed to do: pass on knowledge of competitive Pokémon.

Learning in regards to tiering is what is in question, and that I truly do not see as particularly important.

C&C is about what sets work, what sets are popular. With raw numbers, you can see whether a set becomes popular or not, and it's a fairly simple judgement to decide whether a set works or not through play-testing.

Now, deciding on whether a pokemon should or should not be banned, brings the judgement and human prejudices into the equation much more heavily. It's not just about what works-- it's about what the game should be like.


4 months is a long time my friend, I think I was being plenty reasonable. Even CIM said he could figure out a way to do it in 4 months-- we shall see the fruit of his thread.

 

[capefeather]

I just don't think that it's fair to impose such a restriction of four months when Gen IV UU was fine with a process that could have taken maybe eight. It's great that CIM came up with a shorter process, but even so this should just be a goal, not an absolute demand.

 

[Veedrock]

Not only do quickbans bias against the top legendaries anyway, but also run the very notable risk of not stopping soon enough. I'd say "ban 670+es off the bat go from there" is probably the best way to go, as even some of the "no" votes in the thread are less based on wanting "no bans" and more about not wanting to overban. I'd say that banning the 670+ Pokemon early prevents us from getting into a "ban everything that looks overpowered" mentality that might end with us throwing the baby out with the bathwater.

Gonna ignore the other thread for the time being and bump this suggestion since it has merit. I think it's worth noting the poll in UT isn't a supermajority (and we're all about those, right?). A lot people voted no because they didn't want to vote yes to an unknown banlist. A lot of them also argued about "fast bans" and if they see a list that includes theorized quick ban material they may swing their vote. So I'm going to propose we create a 'real' hypothetical starting banlist and throw up a second poll to see where the community sits after that, and then make our final decision.

 

[Seven Deadly Sins]

Not really. If we demand a supermajority and have at least 2 weeks to test somehow I doubt that we will have more than 67% of people vote for something like dugtrio to get banned !

Come on, man, you know better than that. I'm obviously not saying that we're going to go that far down the list, but there are a fair few BST 600s that are questionable yet not over-the-top, and a quickban mentality that looks to potentially reconstruct the Gen 4 tier list may not give things like Deoxys-D, Lati@s, Manaphy, etc. a fair shake.

 

[Firestorm]

Gonna ignore the other thread for the time being and bump this suggestion since it has merit. I think it's worth noting the poll in UT isn't a supermajority (and we're all about those, right?). A lot people voted no because they didn't want to vote yes to an unknown banlist. A lot of them also argued about "fast bans" and if they see a list that includes theorized quick ban material they may swing their vote. So I'm going to propose we create a 'real' hypothetical starting banlist and throw up a second poll to see where the community sits after that, and then make our final decision.

I believe the supermajority was what we used to ban something. I don't think the "Yes" side is going to hit 67% since it has been at just over half of that for the time the poll has been active. Wasn't the thread to see if people would be vehemently against the idea of a no-ban environment? I didn't see it as "what do people prefer". I saw it as "would this alienate our userbase?". The answer to that question appears to be no. That poll was being used as a poll. Not as a strict decision-maker.

 

[Cathy]

It actually is a supermajority, just not a 2/3 supermajority. Furthermore, as Firestorm points out, what the supermajority has always been for is banning things, so applying that to the poll, the "initial ban list" side would have required the supermajority, with "no initial ban list" being the default. However, it was just a straw poll, and the result is very clear.